Prehistoric ceramic containers were tools used in culinary and ceremonial activity. The archaeological record preserves fragmentary remains of these ceramic tools, challenging the archaeologist to interpret their use and function from potsherds rather than from whole pots. We introduce the two-curvature method for reconstructing ceramic vessel shape and volume from assemblages of potsherds. Each point on a ceramic vessel or sherd has two dimensions of curvature, profile and axial. Profile curvature is sensitive to vessel shape, and axial curvature is sensitive to vessel diameter. Since vessel curvature and sherd curvature are the same, measuring profile and axial curvatures of potsherds provides information on parent-vessel shape and size. The two-curvature method is tested with replicated vessels, and its accuracy for measuring vessel parameters from sherd curvatures is assessed. Vessel parameters are estimated accurately from average-sherd-curvature measurements. Data gathered by using this method, on an archaeological assemblage of Kumeyaay ceramics from southern California, show that Kumeyaay pottery consists of 85–90 percent open-mouth hemispherical bowls and 10–15 percent closed-mouth spherical ollas. Through time, Kumeyaay vessel volume increased while vessel shape remained consistent.

Comments by Dunnell and Yorston (this issue) on our article concerning the effects of tillage on archaeological distributions are, in general, useful contributions toward an understanding of a topic with which most archaeologists must ultimately come to grips. A few of these comments have stimulated a reconsideration of our position on a couple of points, and we hope that this exchange of ideas will clarify the situation for anyone who desires to pursue this research trajectory further.

Accelerator Mass Spectrometry (AMS) dating and faunal analysis have served to reidentify an implement, fashioned from what was believed to be a proboscidean fibula, as a cranium and antler section of a modern moose. This reconciles the discrepancy between the generally accepted date for the extinction of mammoths and mastodons (before 10,000 years ago) with the recession of glacial Lake Agassiz from the site after 8,000 years ago.

The most common method used by archaeologists to identify flaked-stone artifacts that were used by prehistoric people to accomplish some task is to inspect an artifact's edges for macroscopic edge damage. The results of a test of this “no-magnification” approach to microwear analysis indicate that such an approach is likely to produce highly inaccurate and biased data in many or most cases.

Gabriel DeCicco's public relations primer, commissioned by the Public Relations Committee of the Society for American Archaeology and published in American Antiquity, is a useful and comprehensive discussion with a serious limitation. DeCicco's (1988) article is a very good technical guide to generating newspaper-based public relations, but it contains only scant consideration of the fundamental questions of what to say, and why to use public relations. In this postscript to DeCicco's primer, I discuss the importance of recognizing these two questions and articulating responses to them.